Autism is one of the most common developmental disorders, with an occurrence rate of approximately 2-5 per 10,000 in the general population, although true prevalence rates may be as high as 1 per 1000. It is a severe disease with a life long course, and is evident as early as one year of age. Characteristic impairments include difficulty processing social and emotional information, language abnormalities, and repetitive or stereotyped behaviors. Additionally, 75 percent of autistic persons function in the mentally retarded range. The etiology of this disorder is unknown, but evidence for genetic influences is storng. Twin studies shown that monozygotic twins are frequently both affected while dizygotic twins have a lower rate of concordance. Studies of families of autism probands show that there is an elevated risk in sibs of probands relative to the general population. Data from both twin and family studies indicates that the genetic component of autism is the result of multiple genes which interact (epistatic multigenic inheritance). We propose to identify the genetic loci involved in autism. A cohort of 200-250 families with 2 or more affected sibs affected with autism will be ascertained and characterized. The typical family will be 2 affected sibs and both parents. The autistic subjects will be rigorously evaluated to establish the diagnosis of autism. They will also be evaluated with a battery neurocognitive test to potential evaluate sbtypes of autism which may provide markers for genetic heterogeneity. Blood samples will be obtained from members (Affected subjects and their parents and DNA prepared for genetic analysis. In cases where a limited amount of blood can be obtained, permanent cell lines will be prepared. DNA from the family subjects will be genotyped for markers scattered on all chromosomes and the data analyzed by sib-pair methods. The ultimate goal is to identify the chromosomal location of all of the genetic loci contributing to autism with the eventual goal of cloning all of these genes. Unraveling th genetics of autism should provide molecular tools for understanding the pathogenesis of autism.